\begin{table}[ht]
\centering\footnotesize
\caption{We summarize the distribution of predicted probabilities for the plaintiffs' race classifications. Note: the ``None'' category includes plaintiffs for whom the \texttt{wru} algorithm returned predicted probabilities less than 0.5 for every racial/ethnic category.}
\label{tab:app_race_pred}
\begin{tabular}{|l|c|c|c|c|c|}
\hline
Classified & Mean & Mean & Mean & Mean & Mean \\
Plaintiff Race & Pr(White) & Pr(Black) & Pr(Hispanic) & Pr(Asian) & Pr(Other)\\\hline
White (40\%) & 0.750 & 0.086 & 0.039 & 0.013 & 0.113 \\\hline
Black (13\%) & 0.200 & 0.666 & 0.026 & 0.008 & 0.100 \\\hline
Hispanic (12\%) & 0.037 & 0.008 & 0.902 & 0.022 & 0.031 \\\hline
Asian (2\%) & 0.032 & 0.016 & 0.019 & 0.823 & 0.109 \\\hline
Other (1\%) & 0.198 & 0.113 & 0.070 & 0.031 & 0.589 \\\hline
No Dominant Prediction (22\%) & 0.348 & 0.288 & 0.105 & 0.031 & 0.228 \\\hline
None (7\%) & --- & --- & --- & --- & --- \\\hline
\end{tabular}
\end{table}
